Proceedings of ASME Turbo Expo 2016: Turbomachinery Technical Conference and Exposition GT2016 June 13 – 17, 2016, Seoul, South Korea

GT2016-57857

Review of -To-Synthetic (SNG) Production Methods and Modeling of SNG Production in an Entrained-Flow Gasifer

Xijia Lu* and Ting Wang Energy Conversion& Conservation Center University of New Orleans New Orleans, LA 70148, USA

Abstract In this paper, the coal-to-synthetic natural gas (SNG) 1 Introduction technologies have been reviewed. Steam-oxygen gasification, SNG has a large potential market, particularly for countries hydrogasification, and catalytic steam gasification are the three that don't have domestic natural gas (NG) resources and relies major gasification processes used in coal-to-SNG production. on importing expensive foreign Liquefied Nature Gas (LNG). So far, only the steam-oxygen gasification process is Basically any application that currently uses natural gas could commercially proven by installing a catalytic methanation use SNG. In particular, gasification can be used on-site for reactor downstream of the gasification process after syngas is industrial applications to produce SNG, allowing continued produced, cleaned, and shifted to achieve an appropriate H2/CO operation of natural gas equipment, but from a coal source. ratio for methanation reaction. This process is expensive, less Other than producing electricity, industrial use of natural gas efficient, and time consuming. Ideally, it will be more effective was about 28% of the total domestic consumption in United and economic if methanation could be completed in an once- States. A study from the National Energy Technologies through entrained-flow gasifier. Technically, this idea is Laboratory (NETL) in 2007 researched on the feasibility of on- challenging because an effective gasification process is site gasifiers in industrial facilities for the production of SNG typically operated in a high-pressure and high-temperature found that many industrial sites could benefit from the use of condition, which is not favorable for methanation reaction, relatively small gasifier systems to either produce SNG or which is exothermic. To investigate this idea, a computational generate power, H2, or syngas. This is particularly attractive to model is established and a sensitivity study of methanation countries that don't produce NG domestically and have reactions with and without catalysts are conducted in this study. concerns on their national energy security. In modeling the methanation process in a gasifier, correct Several advantages are associated with producing SNG from information of the reaction rates is extremely important. Most for countries without NG resources. SNG can be of known methanation reaction rates are tightly linked to the produced through the gasification of coal and modern gas catalysts used. Since the non-catalytic reaction rates for cleanup system before used for combustion. In this approach, methanation are not known in a gasifer and the issues are SNG is able to substitute for natural gas, a more volatile compounded by the fact that inherent minerals in coal ashes can commodity than coal. In this way, gasification of coals to SNG also affect the methanation kinetics, modeling of methanation helps increase fuel diversity, protecting against an over-reliance in an entrained-flow gasifier becomes very challenging. on a single energy source, and making utilization of coals much Considering these issues, instead of trying to obtain the correct cleaner than the conventional coal burning scheme. Since SNG methnation reaction rate, this study attempts to use in use is identical to natural gas, SNG can be transported and computational model as a convenient tool to investigate the distributed using existing natural gas infrastructure and utilized sensitivity of production under a wide range in existing natural gas–fired power plants or devices such as methanation reaction rates with and without catalysts. From this industrial burners, boilers, kilns, etc. Furthermore, natural gas is sensitivity study, it can be learned that the concept of the fuel that powers most (but not quite all) chemical and implementing direct methanation in a once-through entrained- refining processes, and natural gas is the feedstock for flow gasifier may not be attractive due to competitions of other hydrogen production (for hydro-cracking, hydro-desulfurization, reactions in a high-temperature environment. The production of and ammonia) and syngas production (for methanol, and its SNG is limited to about 18% (vol) with catalytic reaction with a derivatives e.g. MTBE, formaldehyde, and acetic acid). Natural pre-exponential factor A in the order of 107. A further increase gas condensate (ethane and propane) is an advantaged raw of the value of A to 1011 doesn't result in more production of material via ethylene and propylene to much of the organic SNG. This SNG production limit could be caused by the high- chemicals industry (compared to crude-oil-derived naphtha). temperature and short residence time (3- 4 seconds) in the The biomass can also be used along with coal to produce entraind-flow gasifier. SNG. The use of biomass would reduce the ------1 * Xijia Lu is now affiliated with 8 Rivers Capital. emissions, as biomass is a -neutral fuel. In addition, the 1.1 Coal-to-SNG Technology development of SNG technology would also enhance other Steam-oxygen gasification, hydrogasification, and catalytic gasification-based technologies such as hydrogen generation, steam gasification are the three gasification processes used in the Integrated Gasification Combined Cycle (IGCC), or coal-to- coal-to-SNG technology. So far, the proven and liquid technologies, as the production of SNG is at least similar commercialized method of gasification for the coal-to-SNG to these other processes/technologies (Chandel and Williams, process is the steam-oxygen gasification process. A review of 2009 [1]). The addition of a water-gas shift reaction system and these processes is conducted below. methanation reactor to the base gasification and gas cleanup system, as illustrated in Fig.1, allows for the production of SNG 1.1.1 Steam-oxygen gasification that meets pipeline quality natural gas specifications. The In the steam‐oxygen process of converting coal to SNG, coal Reference Plant in Fig. 1 uses a sour-shift reaction to increase is gasified with steam and oxygen. Oxygen is used for partial- the hydrogen concentration in a portion of the syngas from the combustion with char to provide enough energy used for gasifier, so that when this gas stream is mixed with the gasification process. The gasification process produces carbon remainder of the syngas and is cleaned in the Rectisol® AGR monoxide (CO), hydrogen (H2), (CO2), methane system, the cleaned syngas has the proper hydrogen to carbon (CH4), and higher hydrocarbons such as ethane (C2H6) and ratio (3:1) for methanation ( DOE report, 2007 [2]). propane (C3H8). The gas composition depends upon the gasifier's operating conditions, i.e., temperature and pressure. At higher temperatures and pressures, the major products are CO and H2. Three moles of H2 are required to react with each mole of CO to produce one mole of CH4. The concentration of H2 in syngas is increased by a step called the water‐gas shift reaction, which is followed by gas cleaning. The cleaned gas, consisting primarily of CO and H2, reacts in the methanation reactor in the presence of a catalyst to produce CH4 and H2O. The resulting gas, after H2O condensation and polishing, if required, is called synthetic natural gas (SNG). Figure 2 shows the flow diagram of the steam‐oxygen gasification process. The essential components of the process are the air separation unit, the gasifier, the water‐gas shift reactor, the syngas cleanup

Figure 1 Sketch of Major Systems Comprising SNG Production system, and the methanation reactor. Reference Plant (DOE report, 2007 [2])

There are also many challenges associated with the deployment of SNG. In a 2007 DOE/NETL study [2], potential industrial customers of coal-to-SNG gasification for onsite use in NG applications indicated that reliability is important and needs to be near 100%, either through increased performance or redundancy. Some applications are able to also fire oil, allowing for onsite storage of the backup fuel. Availability is still a challenge for gasification, although some sites have Figure 2 The steam-oxygen gasification process diagram (Chen, L., et al., achieved very high availability. The Great Plains Synfuels Plant 2009 [3]) in Beulah, North Dakota, for example, has consistently produced 90- 92% of its rated output capacity [2]. The methanation reaction with catalysts which are mainly In United States, producing SNG from coal is more ruthenium, cobalt, nickel and iron can be described by expensive than producing natural gas through hydraulic CO + 3H2  CH4 + H2O (1) fracturing (fracking) process. For this reason, NETL focuses on As the methanation reactions are highly exothermic and locations and applications where the gasifier could be pressure-favorable, the methanation reactors are designed to run integrated with an industrial process that uses natural gas. This at low temperature and high pressure with catalysts. In the would improve plant economics and would guard the facility methanation reactor, CO and H are converted to CH and H O against fluctuating natural gas prices, because existing coal 2 4 2 in a fixed‐bed catalytic reactor. Since methanation is an transport infrastructure is well-developed, and coal is both exothermic reaction, the increase in temperature is controlled abundant and relatively inexpensive. Another challenge to coal- by recycling the product gas or by using a series of reactors. to-SNG is in transporting a gaseous fuel, which can be difficult Steam is added to the reaction to avoid coke formation in the because of the gases’ low densities. SNG must be cooled and reactor. After the steam is removed from the product gases by then compressed for transport through a close-to-capacity condensation, SNG is ready for commercial applications. A pipeline infrastructure. In addition, pipelines are restricted by conventional design for the methanation process uses three geographical features like oceans, for example. It can be stages. Figure 3 shows the schematic diagram of the ADAM II liquefied (i.e., as Liquefied Natural Gas or LNG) for transport methanation process, illustrating the three-stage design by ships or trucks. documented in Hohlein et al., 1984 [4]. Three adiabatic 2 methanation reactors, D 201, D 202 and D 203 are equipped temperatures and compositions at the inlets and outlets of the with fixed catalytic beds. The syngas coming from the WGS three stages are shown in Table 1. (Chen, L., et al., 2009 [3]) unit is preheated to a temperature above the starting temperature of the catalyst. At each methanation reactor outlet, Table 1 Typical operating conditions and gas compositions in the 3-stage the gas compositions are approximately at chemical equilibrium. methanation process. (Chen, L., et al., 2009 [3]) Heat is generated during the methanation reactions, so syngas cooling is needed between stages. The typical operating

The steam‐oxygen gasification process for SNG has been demonstrated in the Great Plains Synfuel Plant for 20 years and has proven to be successful in practice. The interest in the Coal to SNG concept has grown recently due to the process’s capability for CO2 capture and utilization in enhanced oil recovery. In the Great Plains Synfuel Plant (GPSP), more than 5 million tons of CO have been sequestered up to 2006, which 2 Figure 3 ADAM II 3-stage methanation process at 45 bar, 300-650 ºC doubled the oil recovery rate of an oil field in Saskatchewan. A (Hohlein et al., 1984[4]) detailed process diagram of the GPSP is shown in Fig. 4 (DOE, 2006).

Figure 4 Detailed block flow diagram of the Great Plains Synfuel Plant. (Source: DOE report, April, 2006[5])

The plant consists of a coal and ash handling unit, an Air gas shift reactor, and a methanation unit. Also included are the Separation Unit (ASU), a steam generator, a gasifier, a water AGR plant (Rectisol) and Flue Gas Desulphurization (FGD) 3 unit, which are used to remove the acid gas within the syngas Although this reaction is mildly exothermic, a significant and flue gas, respectively. Besides methane, the plant also amount of energy must be spent in bringing the reactants up to produces ammonia, ammonium sulfate, naphtha, and phenol the operating temperature as well as to sustain the process. (carbolic acid) as by-products. Although demonstrating Methane production is favored at high pressures and the successful and economical clean synthetic fuels production, the process is generally operated at temperatures ranging from 750 GPSP can be further optimized in many aspects. For instance, ºC to 1000 ºC (Higman and Burgt, 2003 [7]). A number of the gasification technology, a Lurgi system, was adopted by the processes were developed and a few of these were operated GPSP 20 years ago and may not be the most favorable option satisfactorily in pilot plant scales. A major issue with today because of its small coal processing throughput and large hydrogasification process is the source of the hydrogen supply production of waste water. Choosing a technology that since hydrogen production can be expensive and hydrogen has produces less waste could eliminate or diminish ancillary a better market value. As natural gas prices have dropped due to processes such as gas liquor separation, wastewater treatment, the recent development of hydraulic fracturing (or fracking), ash handling, and so on. However, replacing the gasification hydrogasification is not attractive, economically. In addition, system would require the adjustment of other processes, the much slower reactivity of carbon with hydrogen compared principally the WGS and methanation systems. to other gasifying agents further hinders the commercialization of hydrogasification. The reactivity of carbon with different 1.1.2 Hydrogasification species at 1073 K and 0.1 atmospheres are shown below 5 The hydrogasification process uses H2 to gasify coal. H2 (Walker et al., 1959 [8]): RO2 (10 )>> RH2O(3) > RCO2 (1) > RH2 -3 reacts with coal to produce CH4. The hydrogasification process (3.1 ) . is exothermic in nature. H2 required for the gasification is either Based on the above discussions, it is evident that, for SNG provided by an external source or by using a methane steam production to be commercially viable, the gasification process reformer. A portion of the CH4 generated in the must solve the two major technical problems faced by hydrogasification reactor is converted into CO and H2 in the conventional hydrogasification and methanation processes. methane steam reformer. The methane These problems are the difficulties in supplying hydrogen in an reaction is: H2O + CH4 → CO + 3H2. 1 mole of CH4 produced inexpensive and simple manner, and also the low carbon from hydrogasification process could generate 3 moles of H2 in conversion ratios observed during conventional the methane steam reformer. During the hydrogasification hydrogasification based processes. process, C +2H2  CH4, only 2 moles of H2 are needed for producing one mole of CH4. That is the reason why H2 required 1.1.3 Hydromethanation (Catalytic steam gasification) for hydrogasification can be provided by using a methane steam Catalysts can be used to enhance the reactions involved in reformer. If CO is shifted to CO2 and H2, the above process will gasification. Many gasifiers must operate at high temperatures lead to a net production of one mole of CH4 and CO2 from two so that the gasification reactions will proceed at reasonable moles of fixed carbon: 2C+2H2OCH4+CO2. A diagram of the rates. Catalysts can also be used to favor or suppress the hydrogasification process is shown in Fig. 5. The formation of certain components in the syngas product. The hydrogasification process is still in the research stage and is not primary constituents of syngas are hydrogen (H2) and CO, but yet commercialized, although a few studies on the process were other products like methane are formed in small amounts. conducted as early as from the 1970s to the 1990s. Ruby et al. Catalytic gasification can be used to either promote methane (2008) [6] proposed a hydrogasification process which consists formation, or suppress it. Disadvantages of catalytic of a hydrogasification reactor, desulfurization and carbonizer gasification include increased materials costs for the catalyst reactors for CO2 removal, and a methanation reactor. itself, as well as diminishing catalyst performance over time. Catalysts can be recycled, but their performance tends to diminish with age. The relative difficulty in reclaiming and recycling the catalyst can also be a disadvantage. In the hydromethanation process, gasification and methanation occur in the same reactor in the presence of a catalyst. Steam is the only gasification agent used so that the water-gas shift. The net reaction route is:

2C + 2H2O  CH4 +CO2 (3)

However, with steam, low temperatures greatly limit the rate Figure 5 Hydrogasification process diagram (Chandel, M., and Williams, of reaction. At high temperatures, the thermodynamic E., 2009 [1]) environment is not favorable for methane production. Therefore, introducing a catalyst at low temperature to facilitate the Hydrogasification was originally developed in the early reaction is highly needed. Alkali metals catalyze carbon with 1900s and there was a revived interest in the process during the steam to form CO and H2 and, by doing so, increase the 1970s and 80s as a result of increasing natural gas prices. The reaction rate several fold. Selection of the catalyst is based on basic reaction is the direct methanation of carbon, as shown its affinity to reacting with coal. KCl and K2SO4, for example, below. are ineffective despite their belonging to the alkali family C +2H2  CH4 H1000K = -89.9kJ/mol (2) (Probstein and Hicks, 1982 [9]). In the process of gasification, the actual catalyst is not retained in the gasifier, but is carried 4 out with the ash. In order for a commercial plant to maximize serves as a feedstock calibration facility for designing profits, it is essential that a recycling loop be implemented to commercial plants (GreatPoint Energy, 2013[13]). recover the catalyst for re-use in the coal gasification process. From end-to-end, in hydromethanation, coal is pulverized and 1.2 Use of biomass for SNG mixed with the selected catalyst. Before feeding the SNG produced from biomass, also known as “bio‐SNG,” has impregnated catalyst-coal into the gasifier, it is dried to remove the advantage of being carbon‐neutral, and, in conjunction with as much moisture from the fuel as possible. Gasifiers are then CO2 capture, the entire process could generate negative carbon fed with the feedstock and begin introducing steam into the emissions. The challenges of using biomass arise due to the environment to perform the gasification. Beyond the chemical composition of biomass, lower calorific value, higher hydromethanation process, carbon monoxide and hydrogen moisture content, and tar formation. The seasonal variation in must be separated from the methane product. A cryogenic the biomass supply and moisture content could require large distillation process effectively separates methane from the amounts of storage space and large drying capacities for synthesis gas (a process with an energy penalty lower than the commercial‐scale biomass gasification units. Another possible oxygen separation from air in an ASU). way of utilizing biomass would be in a coal-biomass co- The advantages of hydrogasification and hydromethanation gasification process. Co-gasification could make it possible to are that they do not use direct combustion to provide heat, so an install large-scale gasification plants, which could be more air separation unit is not required to provide oxygen. Hence, commercially viable. Fluidized bed gasifiers may be better there is less of an energy penalty for the process. Furthermore, suited for biomass gasification than entrained-flow gasifiers as the costs are lower, as the gasification and methanation occur at they can handle variations in size, density, moisture, and tar a lower temperature. The disadvantages of hydromethanation formation. are the production of tar and the separation of the catalyst from In Meijden’s (2010 [14]) study, he categorized the the ash/slag and the loss of reactivity of the catalyst. A diagram gasification technologies associated with bio-SNG process in of the hydromethanation process is shown in Fig. 6. different ways. Based on the gasifier’s type, they are: 1) entrained flow, 2) fluidized bed, and 3) fixed bed. Fluidized bed gasifiers can be divided into two main categories: Bubbling Fluidized Bed (BFB) and Circulating Fluidized Bed (CFB). A bubbling bed is the classical approach where the gas at low velocities is used and fluidization of the solids is relatively stationary, with some finer particles being entrained. At higher gas velocities, a circulation of the bed material is required. This type of gasifier is called a Circulating Fluidized Bed (CFB) gasifier. The typical fluidization velocity in the circulating gasifier is normally between 3 and 10 m/s. The bed material and recycling char are removed from the product gases by a

Figure 6 A diagram of hydromethanation process (Chandel, M., and cyclone or another separation device. Those particles are Williams, E., 2009[10]) recycled back to the gasifier via a non-mechanical valve. The gasification technologies can be implemented with a direct The hydromethanation process developed by Great Point heating scheme or an indirect gasification/heating scheme. For Energy Inc. is considered to be a great advancement in SNG indirect gasification/heating, the conversion of the fuel is being technology. This process was initially developed by Exxon in done in two separate reactors (indirect twin beds). The first the 1970s using potassium carbonate (K2CO3). However, the reactor is for combustion to generate heat for the gasification current Great Point Energy process involves a single reactor process in the second reactor. The char and bed material (e.g. using a proprietary, recyclable catalyst developed in-house and sand) are fed to the combustion reactor. The char is combusted made from abundant, low-cost metals. The catalyst was to produce the required heat for the gasification reactor. The developed with the help of Southern Illinois University, the bed material (sand) carrying the required heat is then University of Toronto, and the University of Tennessee. The transported into the gasification reactor. The biomass in the heat released in the SNG process is sufficient to sustain the gasification reactor is converted into producer (or product) gas gasification, eliminating the need to fire up the reactions with and char (pyrolysis). Char and bed material are separated from purified oxygen. The process was demonstrated with a the gas and returned into the combustion chamber by a solid weeklong pilot run in November 2007. The pilot plant for the gas separation device, such as a cyclone. The producer gas exits process is a 60-foot-high gasifier with an internal diameter of the gasifier and is sent to the gas cleanup system. 14 inches. The price of pipeline-quality natural gas by Great In Meijden’s (2010) [14] study, he used Aspen Plus to Point Energy’s process could be less than $3 per MMBtu simulate a large scale SNG system with 1 GW (HHV) of input (Fairley 2007). Great Point Energy Inc. and Peabody were power. The net overall efficiency on the LHV basis, including working together to commercialize the technology with the goal electricity consumption and pretreatment, but excluding of developing a coal to SNG plant at or near Wyoming’s transport of biomass, is 54% for the BFB, 58% for the CFB Powder River Basin area (GreatPoint Energy 2008[11] [12]). with direct heating/gasification, and 67% for the CFB with the The Company currently conducts tests at a pilot plant at the indirect heating/gasification technique. Energy and Environmental Research Center in Grand Forks, In lieu of the high efficiency of indirect gasification, the ND, which demonstrates the latest version of its technology and Energy Research Centre of the Netherlands (ECN) has demonstrated SNG generation from biomass (Mozaffarian et al. 5

2003, 2004[15] [16]) using the indirect gasification technology 1.3 Recent research on SNG at atmospheric pressure. The process is shown in Fig.7. The Recently, The Arizona Public Service Company (APS) biomass is gasified in the riser of a gasification reactor and the along with the Department of Energy and other partners are remaining char is circulated to the combustor. In this process, developing a hydrogasification process to co-produce SNG and the heat required for gasification is supplied by char electricity from western coals. The objective of the $12.9 combustion in the combustor. Steam is used for gasification and million project is to develop and demonstrate an engineering air is used for char combustion. The lab scale gasifier, scale hydrogasification process which can produce SNG at a developed in 2004, has a biomass capacity of 5 kg/h and cost of less than $5/MMBtu and can utilize low rank western operates at temperatures of 750ºC to 900ºC (Zwart et al. 2006 coal (NETL, 2008 [18]). The Western Research Institute (WRI) [17]). Direct heating/gasification was also tested, which uses is working on the development of a gasification process which oxygen and steam for gasification via a bubbling fluidized bed uses counter-current cyclonic methods in a unique sequence and operates at 850ºC. The gas treatment in the integrated bio- that causes activated carbon char to react with synthesis gas, SNG system consists of tar removal with organic scrubbing both derived from coal. The method does not require pure liquid technology, and and HCl removal with adsorbents. oxygen to produce the synthesis gas (WRI, 2008[19]). KBR developed a new KBR TRIG gasification-based coal- to-SNG. The process shown in Fig. 8 is well suited for a wide range of feedstocks, particularly low-rank coals that are low- cost and abundant. The process scheme offers a technically robust and energy efficient design, with several advantages over comparable gasification processes. The economics of building mine-mouth 150,000 standard cubic feet per day coal- to-SNG facilities using KBR’s TRIG gasification technology is currently being investigated for various western U.S. locations (Ariyapadi et al., 2008 [20]). Figure 8 depicts a simplified block flow diagram illustrating the connectivity between major process units of the KBR system. A cluster of three TRIG gasifiers supply the necessary syngas feed with the appropriate H2: CO ratio to the methanation unit. The main process units include gasification, shift, COS hydrolysis, ammonia scrubbing, mercury removal, acid gas removal, sulfur removal, CO2 compression, methanation, and SNG drying and compression. The detail of each section is described by Ariyapadi et al. (2008) [20].

Figure 7 Simplified scheme of MILENA biomass gasification process (C.M. van der Meijden, 2010[14])

Based on the experiments, the SNG system consists of an indirect gasifier, the so-called MILENA gasifier (Fig. 7), a tar removal system which recycles tar to the gasifier, a gas cleaning and WGS reactor, and a methanation combined reactor. The gasifier contains separate sections for gasification and combustion. The gasification section consists of three parts: the gasifier riser, settling chamber, and downcomer. The red arrows in Fig. 7 represent the circulating bed material. The gasifier working at 850ºC produces nearly -free syngas and a high amount of methane. Tar is recycled to the gasifier in order to increase efficiency, whereas the tar-free syngas is cleaned from other contaminants (e.g., sulfur and chlorine). The clean syngas is fed to a combined shift and methanation process, converting the syngas into SNG. After methanation, further upgrading (e.g., CO2 and H2O removal) is required in order to comply with the desired SNG specifications. The Figure 8 Block Flow Diagram of KBR TRIG Coal-to-SNG Process overall net thermal efficiency is reported as 70% by Low (Ariyapadi et al. 2008 [20]) Heating Value (LHV) basis (approximately 64% HHV basis). Forty percent of the carbon of the biomass becomes part of the A new SNG production technology, called the steam hydrogasification reactor (SHR), which is based on a SNG and an equal amount of carbon is captured as CO2. The remaining 20% of the carbon in biomass becomes flue gas from combination of the hydrogasification and steam pyrolysis the process. reactions, is newly developed by the University of California, Riverside. The configuration of this process allows the use of recycled hydrogen as feed, thus eliminating the hydrogen 6 supply problem. This steam hydrogasification process generates an SHR gasifier contains considerable amounts of methane. a product gas stream with high methane content. The The concentration of methane increases with decreasing composition of the product gas from steam hydrogasification H2O/Feed mass ratio and increasing H2/C feed mole ratio. can be controlled by varying the steam to carbon and H2 to Operating at higher pressures also favors an increase in carbon ratios of the feed. Methane concentration of the SHR methane production. product gas can be varied from 10 to 30 % on a molar basis. Tunå (2008) [22] evaluated twelve different systems for The product gas also contains CO, CO2, H2, and a considerable production of SNG by using Aspen Plus. The system consists amount of unreacted steam. In the SHR gasifier, the feed is of three gasifiers: an entrained-flow, fluidiszd-bed, and indirect transported into the reactor via a slurry. The slurry feed gasifier. Both an isothermal methanation process and an eliminates the need for cumbersome reactor feed systems such adiabatic methanation process have been modeled. Gas cleanup as a lock hopper. This also simplifies feedstock processing was performed using both conventional zinc oxide since drying the feed is not necessary. A portion of the desulfurization with PSA upgrade and a Rectisol® wash. The necessary steam enters the reactor as liquid water that is part of simulation results show that SNG efficiencies from biomass to the slurry and the rest of the steam is superheated and fed along methane of 50% are possible with either gasifier. The fluidized- with the hydrogen. Steam hydrogasification of carbonaceous bed and indirect gasifiers were able to produce SNG with an feedstocks results in improved carbon conversion compared to efficiency around 67%. Furthermore, utilizing a Rectisol gas hydrogasification. An SHR also generates a product gas with a cleanup system does not have a significant negative impact on considerable amount of methane compared to conventional SNG efficiency, but it affects overall efficiency. The simplest partial oxidation gasifiers. The steam hydrogasification reactor system—zinc oxide desulphurization with PSA gas cleanup— can be coupled with a shift reactor, resulting in a gasifier coupled with either methanation system is considered by Tuna configuration that generates a syngas with high methane as the most promising choice. It is based on well-established, concentrations. This configuration also allows considerable widely-used equipment and it offers better efficiency than a control over the final product gas composition. Figure 9 shows wet-gas cleanup process such as Rectisol. If there is a the process configuration involving SHR gasification to significant amount of sulfur in the gas stream that needs to be produce syngas with a high methane content. The slurry made removed, or if the carbon dioxide needs to be captured and of the carbonaceous feed (coal) and water, along with the removed, the Rectisol method will become a competitive option. recycled hydrogen are fed to the SHR, operating at Typically, carbon dioxide capture is not necessary for biomass- approximately 850 ºC and 400 psi. based plants as the carbon emissions are considered neutral. Chen et al. (2009) [23] reviewed the state-of-the-art technologies for Coal-to-SNG, conducted a thermodynamic parametric study of the main components in this process, and and also made an efficiency assessment of the overall energy system, implementing different gasification technologies, including the hydromethanation process. Their results show the O2/Carbon ratio to be about 0.25 - 0.3 and the H2O/Carbon ratio to be about 1.5 – 2, which are favorable ranges to produce a CH4-rich syngas with a high H2/CO ratio. Higher pressure is favorable to the hydromethanation reaction and increases methane yield. The analysis shows that moving-bed, dry ash Figure 9. Schematic diagram of a steam hydrogasification reactor (SHR) gasification achieves a higher energy conversion efficiency method to produce high CH4 production (Chan and Norbeck, 2009 [21]) (67%) than entrained flow gasification (57%) for the overall Coal-to-SNG process. Hydromethanation is a promising route The SHR generates a high methane content product gas that with about 70% energy efficiency. However, it is still under is subjected to warm gas cleanup in order to remove development because of the challenges for separating the contaminants such as sulfur. The gas cleanup must be catalyst from the ash/slag and recovering the loss of reactivity performed at a temperature above the dew point of water. This of the catalyst. will allow the unreacted steam from the SHR to be directly fed Chandel and Williams (2009) [24] examined the different into the shift reactor along with the product gas. In this case, the technologies for producing SNG, as well as the production shift reactor will be operated as a ‘sour-shift’ reactor with a costs and the environmental impacts of SNG. Their paper sulfur tolerant catalyst. In the shift reactor, the CO present in identified the conditions under which SNG production could be the clean product gas reacts with the steam to produce H2. economically viable. In a low‐carbon economy, the Methane is inert in the shift reactor. Alternatively, the product development of the carbon capture and storage would be one of gas will be cooled down and H2 can be recycled to the SHR as the critical factors in the future development of SNG. In the feed. The recycled hydrogen stream eliminates the hydrogen absence of carbon capture and storage and carbon allowance supply problem. The final product gas in either case contains a price in the future, the SNG could be expensive and may not be high quantity of methane. The experimental results of the steam economically viable. Higher natural gas prices and the selling hydrogasification of coal and wood mixtures in a batch reactor of CO2 to enhance oil recovery could make SNGs economically are presented by Chan and Norbeck (2009) [21]). Their results viable. The levelized cost of producing SNG is $8.42/MMBtu show that the carbon conversion values at 700 ºC were for plants using bituminous coal and $9.53/MMBtu for those approximately 60%, whereas, at 800 ºC, the values were closer using sub‐bituminous coal. With CO2 sequestration, SNG costs to 80%. Their simulation results show that the product gas from would increase to $9.15/MMBtu for bituminous coals and 7

$10.55/MMBtu for sub‐bituminous coals. They also examined high temperature methanation sintering and structure sensitivity the cost of producing Bio‐SNG and they reported that, for by doing experiments and found that high temperature keeping the bio‐SNG price lower than $12/MMBtu, the methanation plays a role in the manufacture of SNGs. The key biomass price should not exceed $2.2/MMBtu. The cost of problem is resistance to sintering, which results in a decrease of producing SNG ($8.42-9.53/MMBtu) provided by the analysis both the metal surface area and the specific activity. Paraskevi, of Chandel and Williams (2009) [24] is much more expensive et al. (2008) [32] investigated the catalytic performance of than that ($5/MMBtu) taken from the NETL's report (2008) Al2O3-supported noble metal catalysts for the methanation of mentioned earlier. This further exemplifies the uncertainty in CO, CO2, and their mixture with respect to the nature of the evaluating the true production cost of SNG. dispersed metallic phase (Ru, Rh, Pt, Pd). Results show that the catalytic performance, apparent activation energy, and 1.4 Methanation reaction rates selectivity of reaction products for the solo- or co-methanation The tail end of an SNG plant must necessarily employ a of CO/CO2 depend strongly on the nature of the metallic phase. catalytic methanation step in order to upgrade the heating value Generally, methanation activity is much higher for Ruthenium to approximately 950 Btu/SCF. This step typically involves the and Rhodium catalysts, compared to Palladium or Platinum, methanation reaction shown in Equation 8.1, which is which tend to enhance the WGS reaction. For the simulations in the other chapters of this dissertation, accompanied by a relatively high heat of reaction (HR = 49.3 kcal/mol). Although the methanation of trace quantities of CO the methanation reactions have been excluded due to low has been practiced commercially for many years in ammonia methane production in the previously studied gasification plants, SNG methanation from coal poses a more severe process. However, for this chapter, since it has improved, the problem due to the high concentrations of CO in the synthesis simulation will focus on modeling the methanation reactions in gas. With nickel methanation catalysts, the reaction rates are the coal gasification process. relatively high, and, consequently, heat is also liberated at very For simulating methanation, Watanabe and Otaka (2006)[33] high rates. Problems connected with localized coking and performed a numerical simulation with the coal gasification catalyst sintering generally lead to reactor design concepts model on the Japanese 2 tons/day, research-scale coal gasifier which employ high recycle ratios as first suggested by Dent et supported by the Central Research Institute of Electric Power al. (1948) [25]. Since the catalyst is always in contact with a Industry (CRIEPI). The rate constants of the methanation -14 4 reacting gas mixture, which contains all or most of the five reaction that they used were A = 5.12 x 10 and E = 2.73x 10 J/kmol for the forward reaction rate and A = 4.4 x 1011 and E = components involved in methanation synthesis: H2, CO, CO2, 1.68 x 108 J/kmol for the backward reaction rate. The influence H2O, and CH4, Saletore and Thomson (1977) [26] decided to conduct an experimental study to determine the methanation of the air ratio on gasification performance, gas temperature reaction rates for synthesis feeds containing all five components. distribution, and product gas composition were presented and They also investigated the effect of steam’s high partial discussed in their paper. NETL (2007) [34] only included the pressures on the methanation reaction rate. This was motivated forward methanation reaction in the coal gasification model. -14 by the fact that steam may be added to the synthesis gas in The constants they used were A = 5.12 x 10 and E = 2.73x 4 order to inhibit carbon deposition and at least one methanation 10 J/kmol. reactor’s design concept utilizes a large excess of steam ( Dent et al., 1948) [25]. 2. Objective and Goals Early work on methanation kinetics was accomplished at the The above review shows that steam-oxygen gasification, University of Michigan (Akers and White, 1948) utilizing a 3.2 hydrogasification, and catalytic steam gasification are the three mm commercial nickel catalyst. They correlated their results by major gasification processes used in coal-to-SNG production. assuming that the rate-determining steps were surface reactions, So far, only the steam-oxygen gasification process is although there was some evidence of strong pore diffusion commercially proven by installing a catalytic methanation reactor downstream of the gasification process after syngas is effects. utilized CO2 in place of CO and found that the rate of produced, cleaned, and shifted to achieve an appropriate H2/CO CO2 methanation, CO2 +4H2  CH4 + 2H2O (HR = 165 MJ/kmol), was two orders of magnitude less than the CO ratio for methanation reaction. This process is expensive and methanation rate. Schoubye (1969,) [27] employed small- not effective. Ideally, it will be more effective and economic if sized nickel catalysts at high pressures and concluded that the methanation can be completed in an once-through gasifier (or reaction order with respect to CO was -0.5 at high CO called direct methanation). Technically, this idea is challenging concentration (over 20%) and that the data was best correlated because an effective gasification process is typically operated in a high-pressure and high-temperature condition, which is not by assuming that H2 adsorption determined the reaction rate. Negative reaction orders with respect to CO were also found by favorable for methanation reaction. In the hydromethonation Betta et al. (1974) [28] and Vannice (1975) [29], although (catalytic steam methanation) technology, GreatPoint Energy they all worked at pressures of 1 atm or less. Saletore and invented proprietary catalysts that can catalyze methanation in Thomson (1977) [30] conducted the measurements of the relatively low-temperature environment in a fluidized-bed methanation reaction rate with a 1.6-mm nickel catalyst gasifier. The issue associated with low-temperature catalytic utilizing feed compositions typical of recycle reaction reaction is the forming and removal of tar, which will have an adverse impact on the production rate of methane. But due to configurations and product streams with a high CO2 content. The apparent reaction orders for hydrogen and steam were proprietary nature of the technology, not much technical found to be 0.85 and -0.9 respectively, but there was no information has been released in open literature. Furthermore, significant dependency of the methanation rate on the carbon the throughput (or the yield rate) of methane from a fluidized- oxides. Rostrup-Nielsen, et al. (2007) [31] investigated the bed gasifier is lower than that of an entrained-flow bed. 8

Therefore, one of the objectives of this study is to investigate range methanation reaction rates from non-catalytic condition the methanation process in an entrained-flow gasifier instead of with A in the order of 10-14 to catalytic condition with A in the a fluidized-bed gasifier at higher-temperature environment order of 1011. without tar formation. Since methanation is a reversible catalytic reaction, most of 3. Global Gasification Chemical Reactions the reaction rates for the methanation reaction were obtained This study deals with the global chemical reactions of coal from experiments with specific catalysts under laboratory gasification that can be generalized in reactions (R1.1) through conditions of relatively narrow ranges of pressure and (R1.11) in Table 2. temperature. However, the pressure and temperature conditions The volatiles are modeled with a two-step thermal cracking are very different than the operating conditions in an entrained- process (R 1.8) and gasification processes (R 1.9 and 1.10) with flow coal gasifier. Therefore, it is not clear how the published CH4 and C2H2 as intermediate by-products. The coal used in reaction rates can be trustfully used to predict the actual this study is Illinois No.6 coal, whose composition is given in methanation reaction rate in a gasifier without the presence of Table 3. The compositions of the volatiles are derived from the catalysts and under different temperature and pressure coal’s heating value, proximal analysis, and ultimate analysis. conditions than those used in the laboratory. This challenge is The oxidant is considered to be a continuous flow and the further compounded by the fact that inherent minerals in coal coal particles are considered to be a discrete phase. The ashes can also affect the methanation kinetics. Therefore, due volatiles are modeled to be thermally cracked to CO, H2, CH4, to the unavailability of appropriate methanation reaction rates and C2H2. The N, Cl, and S components are assumed to be for broad operating conditions with different types of coals in converted to N2, HCl, and H2S and COS, respectively. Based on actual gasifiers without using catalysts, the primary objective of the DOE/NETL report (2011) [39], the mole ratio of H2S/COS this study is to use computational model as a convenient tool to is modeled as 5:1. All of the products from cracking the investigate the sensitivity of methane production under a wide volatiles are considered to be a continuous gas phase.

Table 2 Summary of reaction rate constants used in this study

n Reaction k = AT exp(-E/RT) (n=0) Reactions Reaction Type heat,H°R Reference (MJ/kmol) A E(J/kmol) Heterogeneous Reactions

7 R 1.1 C(s) + ½ O2 → CO Partial combustion -110.5 0.052 6.110 Gasification, Chen et al.(2000)[36, R 1.2 C(s) + CO → 2CO +172.0 0.0732 1.125108 2 Boudouard reaction 37] 8 R 1.3 C(s) + H2O → CO + H2 Gasification +131.4 0.0782 1.1510

R 1.4 C +2H2  CH4 Hydrogasification +89.9 N/A Homogeneous Reactions Westbrook and Dryer R 1.5 CO + ½ O → CO Combustion -283.1 2.21012 1.67108 2 2 (1981)[38] 10 7 R 1.6 CO+H2O(g)CO2+H2 shift -41.0 2.7510 8.3810 -14 4 Jones and Lindstedt kf = 5.1210 2.7310 (1998)[35] R 1.7 CO + 3H2  CH4 + H2O Methanation -205.7 11 8 kb = 4.410 1.6810 CH2.761O0.264N0.055S0.048Cl0.005 →0.256CO+0.466H +0.33 2 Two-step Volatiles R 1.8 CH +0.2C H +0.0275N + +4.75 4 2 2 2 Cracking 0.005HCl+0.04H2S Eddy dissipation +0.008COS N/A

Volatiles gasifi- R 1.9 CH4 + ½O2 → CO+2H2 - 35.71 cation via CH4 R Volatiles gasifi- C2H2 + O2 → 2CO + H2 -447.83 1.10 cation via C2H2 Jones and Lindstedt R1.11 H + ½ O → H O Oxidation -242 6.8x1015 1.68x108 2 2 2 (1998)[35]

1) All H°R at 298K and 1 atm. 2) “+” Endothermic (absorbing heat), “-” Exothermic (releasing heat)

9

model considers the chemical transformation of the coal Table 3 The proximate and ultimate analyses of Illinois structure during devolatilization. It models the coal structure No.6 bituminous coal transformation as a transformation of a chemical bridge network, which results in the release of light gases, char, and tar Ultimate Analysis (wt%) Proximate Analysis (wt%) [40]. The initial fraction of the bridges in the coal lattice is 1, Moisture 11.12 Moisture 11.12 and the initial fraction of char is 0. The lattice coordination VM 34.99 Ash 9.7 number is 5. The cluster molecular weight is 400, and the side Ash 9.7 C 63.75 chain molecular weight is 50. The Chemical Percolation H 4.5 Devolatilization (CPD) model is used as the devolatilization Fixed Carbon 44.19 N 1.25 model. Heating value 27.1 S 2.51 (MJ/kg) O 6.88 4.2 Coal or liquid particle motion theory Cl 0.29 In this study, coal and water particles are treated as discrete phases, so the Lagrangian method is adopted to track each particle. The discrete phase is justified in entrained-flow 4. COMPUTATIONAL MODEL gasification process because the average particle concentration The computational model and submodels (devolatilization, is lower than 10%. Particles in the airflow can encounter inertia reactions, particle dynamics, gasification) used in the study and hydrodynamic drag. Because of the forces experienced by follow the methodology developed by the authors' research the particles in a flow field, the particles can be either group in Silaen and Wang [41 and 42] and Lu and Wang [43]. accelerated or decelerated. The velocity change is determined Therefore, the governing and associated equations and detailed by the force balance on the particle, which can be formulated modeling intricacies are not repeated here, but they are briefly by: summarized below. The time-averaged, steady-state Navier- du Stokes equations as well as the mass and energy conservation p  FD  Fg  Fx (4) equations are solved. Species transport equations are solved for dt all gas species involved. The standard k- turbulence model where FD is the drag force per unit particle mass and: with standard wall function is used to provide closure. The P1 18 C Re F  D v - v m (5) model is used as the radiation model. The flow (continuous D 2 24 p p phase) is solved in Eulerian form as a continuum while the pd p particles (dispersed phase) are solved in Lagrangian form as a where mp is the particle mass, dp is the particle diameter, v is discrete phase. A stochastic tracking scheme is employed to the fluid phase velocity, vp is the particle velocity,  is the fluid model the effects of turbulence on the particles. The continuous phase density, p is the particle density, g is gravity,  is the phase and discrete phase are communicated through drag fluid phase molecular viscosity, and CD is the drag coefficient. forces, lift forces, heat transfer, mass transfer, and species The gravitational force, Fg, is calculated as the second term in transfer. equation 4 as:

g p   4.1 Discrete Phase Modeling Fg  m p (6) Gasification or combustion of coal particles undergoes the  p following global processes: (1) inert heating, (2) evaporation of The relative Reynolds number, Re, is defined as: surface moisture, (3) devolatilization and demoisturization, (4) coal d p v p - v Re  (7) combustion and gasification, and (5) ash deposition. The  initially inert coal particles will go through a heating process to increase the particle temperature. When the surface temperature Fx in Eq. 4 is an additional acceleration (force/unit particle mass) term, and typically includes the “virtual mass” force, of a coal particle reaches the vaporization temperature, Tvap, the surface moisture starts to evaporate. Water evaporation thermophoretic force, Brownian force, Saffman's lift force, etc. In this study, the thermophoretic and Saffman’s list forces are continues until the droplet reaches the boiling point, Tbp, when the inherent moisture starts to evaporate and gets driven out. In included. the meantime, devolatilization takes place when the temperature of the coal particle reaches the vaporization 4.2.1 Saffman's lift force temperature of the volatiles, and remains in effect until the The Saffman's lift force, or lift due to shear, is based on the volatiles are completely vaporized out of the coal particles. derivation from Li and Ahmadi [44], which is expressed in a Here, the vaporization temperature refers to combusting generalized form originating from Saffman [45]: 1/ 2    materials (volatiles), and is different from the vaporization 2K di j temperature of surface moisture. Silaen and Wang [41] F  1/4 (v v p ) (8) compared the effect of four different devolatilization models on pdp (dlkdkl ) the gasification process. They concluded that the rate calculated where K = 2.594 and dij is the deformation tensor. This form of by the Kobayashi two-competing rates devolatilization model is the lift force is intended for small particle Reynolds numbers. very slow, while that of the Chemical Percolation Also, the particle Reynolds number based on the particle-fluid Devolatilization (CPD) model gives a more reasonable result. velocity difference (slip velocity) must be smaller than the Therefore, the CPD model was chosen for this study. The CPD square root of the particle. The Reynolds number is based on 10 the shear field. In this study, Saffman's lift force reaches about concentration of the vapor at the particle’s surface, which is 30% of Fg, so it is included in the particle motion model. evaluated by assuming that the flow over the surface is saturated. C is the vapor concentration of the bulk flow, 4.2.2 Thermophoretic Force obtained by solving the transport equations. The values of kc When a particle exists in a flow field with temperature can be calculated from empirical correlations by Ranz and gradients, the force that arises on the particle due to this Marshall (1952) [47], temperature gradient is called the thermophoretic force. This k d force is caused by the unequal momentum between the particle c 0.5 0.33 Sh d   2.0  0.6Red Sc (11) and the fluid. The higher molecular velocities on one side of the D particle due to the higher temperature give rise to more where Sh is the Sherwood number, Sc is the Schmidt number momentum exchange and a resulting force in the direction of (defined as /D), D is the diffusion coefficient of vapor in the decreasing temperature. An extensive review of thermophoresis bulk flow. Red is the Reynolds number, defined as u/d, u is by Talbot et al. [46] indicated that the following equation for the slip velocity between the particle and the gas, and d is the the thermophoretic force, Fx, provides the best fit with particle diameter. experimental data over a wide range of Knudsen numbers: When the particle temperature reaches the boiling point, the following equation can be used to evaluate its evaporation rate: 6d 2C (K  C Kn) 1 T F   p s t (9) x (1  3C Kn)(1  2K  2C Kn) m T x m t p dm λ d 2   0.5 (12) where  πd  (2.0 0.46Red )ln1 cp (T  T) / hfg / cp dt d Kn = Knudsen number = 2λ/dp   λ = mean free path of the fluid where is the heat conductivity of the gas/air, and hfg is the K = k/kp droplet latent heat. cp is the specific heat of the bulk flow. k = fluid thermal conductivity based on translational energy The particle temperature can also be changed due to heat only = (15/4) µR transfer between particles and the continuous phase. The kp = particle thermal conductivity particle’s sensible heat changes depending on the convective CS = 1.17, Ct = 2.18, Cm = 1.14 heat transfer, latent heat (hfg), species reaction heat (Hreac), and mp = particle mass radiation, as shown in the following equation: T = local fluid temperature dT dm dm m c  A h(T - T)  p h  f p H  A    4 T(13)4  µ= fluid viscosity p p dt p  dt fg h dt reac p p R

where the convective heat transfer coefficient (h) can be This expression assumes that the particle is a sphere and that obtained with a similar empirical correlation to Eq. 14: the fluid is an ideal gas. In this study, the local temperature gradient in the flow field is important because of local hd 0.5 0.33 Nu   2.0  0.6Re Pr (14) combustion and gasification reactions between the coal d λ d particles and gas mixture. Therefore, the thermophoretic force where Nu is the Nusselt number, and Pr is the Prandtl number. is considered in this study. Eq. (12) is used for both water droplets and coal particles.

4.3 Liquid particle model 4.4 Particle Reactions When the coal is injected through the injectors, the water The reactions of the particles occur after the devolatilization content in the coal (for dry-feed cases) and the water used for process has finished. The rate of depletion of solid due to a slurry-feed cases is treated as being in the condensed phase (i.e. surface reaction is expressed as: liquid water), which can't be lumped into the continuous phase, so the liquid water is atomized into small droplets. R  AηR (15) Theoretically, evaporation occurs at two stages: (a) when the where temperature is higher than the saturation temperature (based on R = rate of particle surface species depletion (kg/s) the local water vapor concentration,) water evaporates from the A = particle surface area (m2) droplet’s surface, and the evaporation is controlled by the water Y = mass fraction of the solid species on the surface of the vapor partial pressure until 100% relative humidity is achieved; particle and (b) when the boiling temperature (determined by the gas-  = effectiveness factor (dimensionless) water mixture pressure) is reached, water continues to R = rate of particle surface species reaction per unit area evaporate even though the relative humidity reaches 100%. (kg/m2-s) 3 After the moisture is evaporated due to either high temperature pn = bulk concentration of the gas phase species (kg/m ) or low moisture partial pressure, the vapor diffuses into the D = diffusion rate coefficient for reaction main flow and is transported away. The rate of vaporization is k = kinetic reaction rate constant (units vary) governed by the concentration difference between the surface N = apparent order of reaction. and the gas stream, and the corresponding mass change rate of the droplet can be given by: The particle reaction rate, R, is controlled by the diffusion of dm reactant gases (e.g., O2, CO2, H2, and water vapor) from the d 2 (10)  πd kc (Cs  C ) immediate continuous phase in the very cell, where the particle dt is located, to the particle's surface; it can be expressed as where kc is the mass transfer coefficient and Cs is the 11

R = D (p - p ) = kp N (16) n s s Raw Syngas  Pressure: 24atm

 No slip condition at wall where p is the gas concentration surrounding the particle and p n s  Adiabatic walls is the gas concentration at the particle surface, and k is the  Inlet turbulence intensity 10% kinetic reaction rate constant. However, since ps is not known, nd ps is then expressed as ps= pg - (R/D), so the explicit Top view of 2 stage appearance of ps can be removed by plugging ps into the above injectors Coal equation, resulting in the following equation, N  R  R  kpn   (17) Coal Coal  D  9m The kinetic reaction rate constant is usually defined in an Arrhenius form as n E RT k  AT e . (18) Coal The second term in Eq. (17) indicates the diffusion limit Top view of 1st stage condition that controls the surface chemical reaction rate, k. In injectors Eq. (17) R needs to be obtained by iterations. However, for Coal & O2 reaction order N = 1, R can be solved explicitly as Coal & O2 pn[kD/(D+k)], and the rate of particle surface species depletion 2.25m can be expressed as

kD Coal & O2 R  Aηpn . (19) D  k 0.75m For reaction order N = 0, 1.5m 0.75m Coal & O2 R  Aηk . (20) 0.75m

The unit of the rate of depletion of the solid R is kg/s. The Figure 10 Schematic of the two-stage entrained-flow gasifier 2 kinetic reaction rate constant k (kg/m -s) for the solid-gas char reactions are determined by the kinetic reaction rate constants 4.8 Boundary and Inlet Conditions adopted from published literatures as presented in Table 2. The total mass flow rates of the dry coal and the oxidant are 10.5 kg/s and 7.64 kg/s, respectively. The total mass flow rate 4.5 Turbulent Dispersion of Particles of the coal slurry case is 17.5 kg/s. The difference in fuel mass The dispersion of particles due to turbulence in the fluid flow rates is caused by water added for making the coal slurry. phase is predicted by using a stochastic tracking scheme, which The coal/water weight ratio of the coal slurry varies from 60%- is modeled with the eddy lifetime. In this model, each eddy is 40%. Oxidant/coal slurry feed ratio is such that the characterized by the Gaussian-distributed, random velocity stoichiometric ratio remains at 0.3. The stoichiometric ratio is fluctuations u' , v' , w' , and a time scale  e . Therefore, the defined as the percentage of oxidant provided over the particle trajectories are calculated by using the instantaneous theoretical stoichiometric amount needed for complete flow velocity (u) rather than the average velocity ( ). The combustion of carbon. For the dry coal case, N2 (5% of the total velocity fluctuation is then given as: u = + u' and weight of the oxidant) has been injected with O2 to transport the 0.5 coal power into the gasifier.  2  0.5 The oxidant is considered to be a continuous flow, while the u'  ζu'   ζ2k/3 (21)   coal slurry is considered to be a discrete flow. The discrete where  is a normally distributed random number. This velocity phase only includes the fixed carbon and water from the will apply during a characteristic lifetime of the eddy (te), inherent moisture content of the coal (8.25% wt.) and the water calculated from the turbulence kinetic energy and dissipation added to make the slurry. The slurry coal is treated as particles rate. After this time period, the instantaneous velocity will be containing both coal and liquid water. The walls are all set to be updated with a new  value until a full trajectory is obtained. adiabatic and are imposed with the no-slip condition (i.e., zero velocity). The boundary condition of the discrete phase at the 4.6 Computational Models and Assumptions walls is considered to be “reflect,” which means that the The computational domain and elements on the gasifier wall discrete phase elastically rebounds off once reaching the wall. are shown in Fig.10. The grid consists of 1,106,588 The operating pressure inside the gasifier is set at 24 atm. The unstructured tetrahedral cells. The buoyancy force is outlet is set at a constant pressure of 24 bars. From here, the considered. The varying fluid properties, such as density, syngas is considered to be a continuous flow, and the coal and specific heat value, thermal conductivity, absorption char from the injection locations are considered to be discrete coefficient, etc. are calculated for each species as a function of particles. The particle size is uniformly given as spherical temperature and pressure by using a piecewise polynomial droplets with a uniform arithmetic diameter of 50 m. approximation method. The properties of the gas mixture are Although the actual size distribution of the coal particles is non- calculated using a mass weighted average method. uniform, a simulation using uniform particle size provides a more convenient way to track the devolatilization process of coal particles than a non-uniform size distribution. 12

increase of reaction rate with A-value changing from 107 to 5 Results and Discussions 1011 does not produce a notable increase of methane. . This gasification-methanation CFD model of this study is built on the previous gasification model without methanation. Table 5 Sensitivity study of methane production and syngas The previous gasification model had been compared reasonably composition on varying A-value for dry coal cases well with the experimental results from the Japanese CRIEPI gasifier [49]. A 5.1210-14 5.1210-7 5.12 5.12107 5.121011 CO 0.35 0.35 0.31 0.27 0.26 5.1 Jones and Lindstedt's forward and backward methanation reaction rates (dry coal case) CO2 0.14 0.14 0.19 0.23 0.24

The first step in the simulation is to use Jones’s and H2 0.26 0.26 0.10 < 0.01 < 0.01 Lindstedt’s forward and backward methanation reaction rates H O 0.13 0.13 0.20 0.22 0.22 for the dry coal case. The result of the syngas composition and 2 temperature at the exit of the gasifier is shown in Table 4. Due CH4 0.04 0.04 0.11 0.18 0.18 to the high backward rate, reactants cannot be totally consumed C H 0.04 0.04 0.05 0.06 0.06 and no CH is produced at the exit of the gasifier. 2 2 4 Other 0.03 0.03 0.04 0.03 0.03 species Table 4 Exit syngas composition and temperature by using the Jones and Lindstedt’s forward and backward rates (dry coal case) T (K) 2237 2238 2487 2520 2556

-14 Forward rate: A=5.1210 , 5.3 Investigation of the sensitivity of methane production by 4 Dry coal case E= 2.7310 J/kmol varying methanation reaction rates of coal slurry cases Backward rate: A=4.41011, 8 Coal slurry cases have also been studied. Table 6 shows that E= 1.6810 J/kmol the results of the slurry coal cases are similar to the dry coal CO 0.38 cases with increased volume fractions of CH4 and H2O as A- CO2 0.12 value increase, but decreased volume fractions of CO and H . H 0.32 2 2 The syngas composition remains the same when the A value H O 0.11 2 decreases below 5.1210-7 or increases above 5.12107. The CH4 <0.01 only difference between the dry and slurry cases is that the C2H2 0.04 Other species 0.03 concentration of CH4 is higher in the dry case (0.18) than in the Temp (K ) 2142 slurry case (0.11).

5.2 Investigation of the sensitivity of methane production by Table 6 Sensitivity study of methane production and syngas varying methanation reaction rates of the dry coal cases composition on varying A-value for coal slurry cases The first approach is to only consider the forward rate as the representative of the net global rate of the methanation reaction. A 5.1210-14 5.1210-7 5.12 5.12107 5.121011 The sensitivity study is performed by consecutively changing CO 0.26 0.26 0.23 0.20 0.20 -14 the pre-exponential rate constant, A, from 5.1210 to CO2 0.09 0.09 0.11 0.13 0.13 11 5.1210 , while the activation energy is kept the same as the H2 0.17 0.17 0.07 < 0.01 < 0.01 4 original value (E = 2.7310 J/kmol). The magnitude order of A H2O 0.39 0.39 0.45 0.49 0.49 -14 ~ 10 is based on the Jones and Lindstedt's reaction rate, and CH4 0.03 0.03 0.07 0.11 0.11 9 the consecutive increase of the order of A up to 10 is based on C2H2 0.03 0.03 0.03 0.04 0.04 Other the reaction rate over nickel-based catalyst [48]. A further 0.03 0.03 0.04 0.03 0.03 increase of A to 1011 is artificial, with the purpose of species investigating the sensitivity of how much more production of T(K) 1672 1671 1837 1878 1879 methane could be harvested if such a catalyst were found that could achieve the reaction rate with an A value reaching 1011. 6. Conclusions Table 5 shows the result of this sensitivity study in terms of In this paper, the coal-to-synthetic natural gas (SNG) syngas temperature and composition at the gasifier exit. It can technologies have been reviewed. Steam-oxygen gasification, be seen that the increased volume fractions of CH4 (from 0.04 hydrogasification, and hydromethanation (catalytic steam to 0.18) and H2O (from 0.13 to 0.22) at the exit, and the gasification) are the three major gasification processes used in decreased volume fractions of H2 (from 0.26 to 0) and CO coal-to-SNG production. So far, only the steam-oxygen (from 0.35 to 0.26) adequately indicate the faster rate of the gasification process is commercially proven, by installing a reaction: CO + 3 H2  CH4 + H2O as the pre-exponential methanation reactor downstream of the gasification process constant value (A) increases from 5.1210-14 to 5.121011. The after syngas is produced and cleaned. However, such a increased exit temperature from 2237 to 2256 K also methanation process is not considered to be an effective means adequately reflects the exothermic nature of the methanation of synthetic natural gas production due to different reasons process. The syngas composition remains the same when the A including relatively poor efficiency. A more efficient way is to value decreases below 5.1210-7 or increases above 5.12107. produce methanation directly in a once-through gasifier via From this sensitivity study, it can be observed that a further hydromethanation. However, the to-be-commercialized hydromethanation process developed by GreatPoint Energy is 13 using low-temperature catalytic steam methanation in fluidized 12. GreatPoint Energy, 2008b. “GreatPoint Energy announces beds. To avoid tar forming and increase throughput, this study coal supply partnership with Peabody Energy and enters investigates methanation process in a high-pressure and high- into agreement to build natural gas manufacturing facilities temperature entrained-flow gasifier through CFD simulation. in Powder River,” Basin. http://www.greatpointenergy. Since both the non-catalytic and catalytic reaction rates for com/ GPE‐Peabody1‐25‐08.pdf. methanation are not actually known in a complicated and 13. GreatPoint Energy, 2015, hazardous environment such as in an entrained-flow gasifier, a https://www.greatpointenergy.com/developmentstatus.php CFD scheme is used in this study as a convenient tool to 14. C.M. van der Meijden, 2010, “Development of the investigate the sensitivity of methane production on the MILENA Gasification Technology for the Production of variation of the methanation reaction rates. The sensitivity Bio-SNG,” Eindhoven University of Technology Library, study is performed by keeping the activation energy of Jones ISBN: 978-90-386-2363-4. and Lindstedt's rates intact but changing the pre-exponential 15. Mozaffarian, M., and Zwart, R., 2003, “Feasibility of constant value, A, from non-catalytic condition in the order of Biomass/waste Related SNG Production Technologies,” 10-7 to catalytic condition in the order of 1011. Both dry-feed http://www.ecn.nl/publications. and slurry-feed conditions have been considered. The results 16. Mozaffarian, M., Zwart, R., Boerrigter, H., Deurwaarder, show that the syngas composition remains the same when the E., Kersten, S., 2004, “Green Gas as SNG (synthetic -7 A-value decreases below 5.1210 or increases above natural gas) ‐ A Renewable Fuel with Conventional 5.12107. From this sensitivity study, it can be observed that a Quality,”http://www.ecn.nl/ publications. further increase of reaction rate with A-value changing from 17. Zwart, R., Boerrigter, H., Deurwaarder, E., Van der 107 to 1011 does not produce a notable increase of methane Meijden C., and Van Paasen. S., 2006, “Production of above 18% (vol.) for dry-feed cases and 11% for slurry-feed Synthetic Natural Gas (SNG) from Biomass,” cases. This SNG production limit could be caused by the high- http://www.ecn.nl/docs/ library/report/2006/e06018.pdf. temperature and short residence time (3- 4 seconds) in the 18. NETL, 2008, “Gasification Technologies, Development of entrained-flow gasifier. a Hydrogasification Process for CO ‐ production of Substitute Natural Gas (SNG) and Electric Power from References Western Coals Description,” http://www.netl.doe.gov/publications/factsheets/project/Pro 1. Chandel, M., and Williams, E., 2009, “Synthetic Natural j410.pdf. Gas (SNG): Technology, Environmental Implications, and 19. WRI, 2008, “Substitute Natural Gas from Coal, Western Economics,” Climate Change Policy Partnership. Research Institute,” Laramie, WY, USA. http://www. netldev netl.doe.gov. http://www.westernresearch.org/business.aspx?ekfrm=204. 2. DOE/NETL Report, 2007, “Industrial Size Gasification for 20. Ariyapadi, S., Shires, P., Bhargava, M., and Ebbern, D., Syngas, Substitute Natural Gas and Power Production,” 2008, “KBR’s Transport Gasifier (TRIG™) – an Advanced National Energy Technology Laboratory. Gasification Technology for SNG Production from Low- 3. Chen, L., Nolan, R., and Avadhany, S., 2009, Rank Coals,” Presented at the Twenty-fifth Annual "Thermodynamic Analysis of Coal to Synthetic Natural Gas International Pittsburgh Coal Conference September 29 – Process,”http://www.mit.edu/people/leichen/Thermodynami October 02, 2008, Pittsburgh, PA. c_Analysis_of_the_coal_to_SNG_process.pdf 21. Chan,P., and Norbeck, J., 2009, “Production of Methane 4. Hohlein, B.; Niessen, H.; Range, J.; Schiebahn, H. J. R.; Rich Fuel Gas from Coal using Steam Hydrogasification,” Vorwerk, M., “Methane from synthesi s gas and operation Presented at the Twenty-sixth Annual International of high-temperature methanation”. Nuclear Engineering Pittsburgh Coal Conference September 21 – 24, 2009, and Design 1984, 78, (2), 241-250. Pittsburgh, PA. 5. DOE Report, 2006, “Practical Experience Gained during 22. Tunå, Per, 2008, “Substitute Natural Gas from Biomass the First Twenty Years of Operation of the Great Plains Gasification,” Synopsis of Master Thesis, Department of Gasification Plant and Implications for Future Projects,” Chemical Engineering, Lund University, P. O. Box 124, Department of Energy. SE-221 00 Lund, Sweden. 6. Ruby, J., Kramer, S., Hobbs, R., and Bryan B., 2008. 23. Chen, L., Nolan, R., and Avadhany, S., 2009, Substitute Natural Gas from Coal Co‐Production Project – a “Thermodynamic Analysis of Coal to Synthetic Natural Status Report. http://www.nexant.com/docs/Service Gas Process,” /energy_technology/SNG.pdf. http://www.mit.edu/people/leichen/Thermodynamic_Analy 7. Higman, C., and M. van der Burgt, Gasification, Elsevier, sis_of_the_coal_to_SNG_process.pdf 2003. 24. Chandel, M., and Williams, E., 2009, “Synthetic Natural 8. Walker, P., Rusinko, F., Austin, L., 1959, "Gas reactions of Gas (SNG): Technology, Environmental Implications, and carbon," Advances in , vol. 11, pp. 133-221. Economics,” Climate Change Policy Partnership. 9. Probstein, R., Hicks, R., 1982, Synthetic fuels. McGraw- http://www. netldev netl.doe.gov. Hill, New York. 25 Dent, F., Moignard, L., Eastwood. A., Blackturn, W., and 10. Chandel, M., and Williams, E., 2009, “Synthetic Natural Hebden, D., 1948, “An investigation into the Catalytic Gas (SNG): Technology, Environmental Implications, and Synthesis of Methane for Town Gas Manufacture,” Gas Res. Economics,” Climate Change Policy Partnership. Board Comm., Vol. 20, pp. 106-121. http://www. netldev netl.doe.gov. 26. Dilip A. Saletore and Wllllam J. Thomson, 1977, 11. GreatPoint energy, 2008a. http://www.greatpointenergy.com. “Methanation Reaction Rates for Recycle Reactor 14

Compositions,” Ind. Eng. Chern., Process Des. Dev., Vol. Flames,” Combustion Science and Technology, Vol. 27, 16, pp. 70-75. pp.31-43. 27. Schoubye, P., “Methanation of CO on some Ni Catalysis,” 39. DOE/NETL Report, 2011, “Cost and Performance Baseline J.Catal. 14, pp.238-246, 1969. for Fossil Energy Plants Volume 3a: Low Rank Coal to 28. Betta. R. A., Piken, A. G., and Shelef, M., 1974, Electricity: IGCC Cases,” National Energy Technology “Heterogeneous Methanation: Initial Rate of CO Laboratory. Hydrogenation on Supported Ruthenium and Nickel,” 40. Fletcher, T., Kerstein, A., Pugmire, R., and Grant, D., 1990, Journal of Catalysis, Vol. 35, pp. 54-64. "Chemical Percolation Model for Devolatilization. 2. 29. Vannice, M. A, 1975, “The Catalytic Synthesis of Temperature and Heating Rate Effects on Product Yields," Hydrocarbons from H2/CO Mixtures over the Group VIII Energy and Fuels, Vol. 4, pp. 54-60. Metals: II. The Kinetics of the Methanation Reaction over 41. Silaen, A. and Wang,T., 2010 " Effect of Turbulence and Supported Metals,” Journal of Catalysis, Vol. 37, pp. 462- Devolatilization Models on Gasification 472. Simulation," International Journal of Heat and Mass 30. Dilip A. Saletore and Wllllam J. Thomson, 1977, Transfer, vol. 53, pp. 2074-2091 “Methanation Reaction Rates for Recycle Reactor 42. Silaen, A. and Wang, T., 2012," Investigation of Coal Compositions,” Ind. Eng. Chern., Process Des. Dev., Vol. Gasification Process under Various Operating Conditions 16, pp. 70-75. Inside a Two-Stage Entrained Flow Gasifier," ASME 31. J. R. Rostrup-Nielsen, K. Pedersen, and j. Shehestedand Journal of Thermal Science and Engineering Applications, High, 2007 " temperature methanation: Sintering and Vol. 4, 021006/1-11 structure sensitivity," Applied Catalysis A : General, vol. 43. Lu, X. and Wang, T., 2014, "Investigation of Low Rank 330, pp. 134-138. Coal Gasification In a Two-Stage Downdraft Entrained- 32. Paraskevi P., Dimitris K., and Xenophon V., 2008, Flow Gasifier, " International Journal of Clean Coal and “Selective Methanation of CO over Supported Noble Metal Energy, Vol. 3, No. 1, pp. 1-12, Online, February 2014 Catalysts: Effects of the Nature of the Metallic Phase on (http://www.scirp.org/journal/ijcce) Catalytic Performance,” Applied Catalysis A: General, Vol. http://dx.doi.org/10.4236/ijcce.2014.21001. 344, pp. 45–54. 44. Ahmadi, G., Mazaheri, A., Liu, C., and Gamwo, I., 2002, 33. Watanabe, H., and Otaka, M., 2006, “Numerical “Computer Modeling of Flow, Thermal Condition and Ash Simulation of Coal Gasification in Entrained Flow Coal Deposition in a Hot-Gas Filtration Device,” 5th Gasifier,” Fuel, Vol. 85, pp. 1935-1943. International Symposium on Gas Cleaning at High 34. DOE/NETL Report, 2007, “Industrial Size Gasification for Temperatures, Morgantown, WV (US), 09/17/2002-- Syngas, Substitute Natural Gas and Power Production,” 09/20/2002. National Energy Technology Laboratory. 45. Saffman, P., 1965, “The Lift on a Small Sphere in a Slow 35. Jones, W.P., and Lindstedt, R.P., 1998, “Global Reaction Shear Flow,” J. Fluid Mech., Vol. 22, pp. 385-400. Schemes for Hydrocarbon Combustion,” Combustion and 46. Talbot, L., Cheng, R., Schefer, R., and Willis, D., 1980, Flame, Vol. 73, pp. 233-240. “Thermophoresis of Particles in a heated Boundary Layer,” 36. Chen, C., Horio, M., and Kojima, T., 2000(a), “Numerical Journal of Fluid Mechanics, Vol. 101, pp. 737-758. Simulation of Entrained Flow Coal Gasifiers. Part I: 47. Ranz, W.E., and Marshall, W.R., 1952, “Evaporation from Modeling GreatPoint of Coal Gasification in an Entrained Drops Part 1 and Part 2,” Chemical Engineering Progress, Flow Gasifier,” Chemical Engineering Science, Vol. 55, Vol. 48, pp. 173-180. pp. 3861-3874. 48. Sehested, J., Dahl, S., Jacobsen, J., and Rostrup-Nielsen, 37. Chen C., Horio M., and Kojima T., 2000(b), “Numerical J., 2005, "Methanation of CO over Nickel: Mechanism Simulation of Entrained Flow Coal Gasifiers. Part II: and Kinetics at High H2/CO Ratios," J. Phys. Chem. B, Effects of Operating Conditions on Gasifier Performance,” 109 (6), pp 2432–2438 Chemical Engineering Science, Vol. 55, pp. 3875-3883. 49. Lu, X. and Wang, T., 2013, "Water-Gas Shift Modeling in 38. Westbrook, C., and Dryer, F., 1981, “Simplified Reaction Coal Gasification in an Entrained-Flow Gasifier, Part 2: Mechanisms for the Oxidation of Hydrocarbon Fuels in Gasification Application," DOI information: 10.1016/j.fuel.2013.02.024, Fuel, vol. 108, pp. 620-628.

15